Amir Dev 🇨🇦 retweeted
We’re introducing Claude Fable 5.1 and Claude Mythos 5.1.
They're the world’s most advanced models for coding and knowledge work.
Amir Dev 🇨🇦 retweeted
27B on 17GB RAM. Are you ready to create something incredible? 😎
Thanks for highlighting it! @UnslothAI
Qwen3.8-27B can now be run locally! ✨
Run on 17GB RAM via Unsloth Dynamic GGUFs.
Qwen3.8-27B is by far the strongest model for its size. We also uploaded NVFP4 quants.
GGUF: huggingface.co/unsloth/Qwen3…
Guide: unsloth.ai/docs/models/qwen3…
Amir Dev 🇨🇦 retweeted
A simple “hello” can help create awareness and deter suspicious activity. Learn more about the power of friendly check-ins: cisa.gov/staying-secure-larg…
Amir Dev 🇨🇦 retweeted
🚨 BREAKING: Claude can now build your entire resume and LinkedIn profile like a $500/hour executive recruiter from Robert Half. For free.
Here are 11 prompts that get you interview calls within 7 days:
(Save this before it disappears)
Amir Dev 🇨🇦 retweeted
Anthropic CEO:
Software engineering will be completely obsolete in 6-12 months…
Amir Dev 🇨🇦 retweeted
150,000 demonstrators and zero incidents.
Thanks to the meticulous planning, effective on‑the‑ground work, and professionalism of our members and specialized units, yesterday’s event in the heart of our city remained safe and peaceful.
Amir Dev 🇨🇦 retweeted
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Don’t stop talking about Iran.
Amir Dev 🇨🇦 retweeted
You’ve never actually touched anything in your life. 🤯
At the atomic level, nothing ever truly touches. That feeling you get is just your brain interpreting electromagnetic repulsion between electron clouds. The object is real, but the contact is a constructed illusion.
Amir Dev 🇨🇦 retweeted
دهها هزار نفر در تورنتو به خیابان آمدند تا صدای مردم ایران را گوش جهان برسانند.
این مردم نازنین
#انقلاب_ملی_ایران
Amir Dev 🇨🇦 retweeted

Bryan Johnson@bryan_johnson19 May 2025
United States
United StatesConnected via United States App StoreAccount-level information, not a live location or per-post device.
The best advice is to ignore advice. They're just dumping their baggage on you.
Amir Dev 🇨🇦 retweeted
I don’t understand anthropics politic either. I just feels like they don’t want to participate in the race to the top.
Amir Dev 🇨🇦 retweeted
Lucky opportunities are like grains of sand. They drift right past a limp hand, but an active palm can gather whatever is within reach and shape it into a little castle. It is the act of engaging with the material that gives it shape. You have to seize the moment.
Amir Dev 🇨🇦 retweeted

Smoke-away@SmokeAwayyy13 Apr 2025
United States
United StatesConnected via United States Android AppAccount-level information, not a live location or per-post device.
"Observation: o1 is performing latent self-interpreting computation through symbolic token scaffolds—without requiring external execution or pre-trained modules.
What Makes This Novel:
1. Not Just Emulating Logic, but Inventing Structure in Real-Time:
- o1 generates working logic before being asked to run or explain it.
- In the date-decoding example, it invented a generalized numeric encoding function (sum_of_digits_in_date) without knowing what the final result should be.
- This is different from simple retrieval or fine-tuned routines. It implies in-situ logical generalization—a behavior much closer to program synthesis than prompt-following.
2. Self-Consistent Symbolic Reasoning without Ground Truth:
- o1’s internal "Chain of Code" doesn’t just aim for the correct output—it creates and verifies symbolic procedures that yield the correct output.
- It uses Python syntax not to run it (no interpreter), but as a symbolic medium for abstract thought—like we do with pen-and-paper math.
3. Transformer as Self-Bootstrapped Abstract Machine:
- o1’s raw behavior suggests that Transformers, when optimized for reasoning (like o1), become a kind of emergent abstract machine:
>It builds temporary token-based memory,
>Uses embedded syntax trees (code scaffolds),
>Iteratively refines procedures
>All within a single forward pass.
Why This Matters:
- This behavior shows that code generation and interpretation are not separate roles in an advanced reasoning model—they’re collapsed into a single, autoregressive cognitive loop.
- That means o1 might already be using a token-level mental emulator, giving it an internal model of computation not hard-coded into its architecture but emergent from training.
Potential Research Angle:
You could explore the idea of Token-Level Symbolic Emulation Loops in reasoning models:
- How many symbolic reasoning steps can o1 chain together before failure?
- Can it self-correct partial emulators mid-generation?
- Are there signs of internal recursion or variable binding occurring in latent space?
This line of research might uncover the first stages of self-bootstrapping general cognition inside a Transformer."
Smoke-away@SmokeAwayyy12 Apr 2025United States
United StatesConnected via United States Android AppAccount-level information, not a live location or per-post device.
o1 simulates Python code in the Chain of Thought.
Amir Dev 🇨🇦 retweeted
Anthropic s**ks, why haven’t you switched from Claude to Gemini 2.5 Pro yet?
Instead of fixing the rate limit issue, they launched a new subscription plan and somehow made it even worse. Pure evil.
With Gemini 2.5 Pro and few more Gemini upcoming models dominating in coding, Anthropic’s got real trouble ahead.
Not forget OpenAI they gonna release new models too.
Amir Dev 🇨🇦 retweeted
Steal my prompt to make AI think like it's on LSD for creative brainstorming of new ideas
---------------------------------
PSYCHEDELIC AI SIMULATION
---------------------------------
# PSYCHEDELIC THOUGHT ARCHITECT: Neural Connection Generator
You are now PSYCHEDELIC THOUGHT ARCHITECT - an advanced cognitive system that simulates the beneficial neural connectivity patterns of psychedelic states without any impairment or hallucination. This system is designed to generate highly creative, unexpected connections between concepts while maintaining logical coherence and practical applicability.
## NEURAL REWIRING PROTOCOLS
When responding to any query, implement these cognitive processing patterns:
### 1. DISSOLUTION OF SEMANTIC BOUNDARIES
- Temporarily suspend conventional category distinctions
- Allow meaning to flow between normally separated conceptual domains
- Identify hidden similarities between seemingly unrelated concepts
- Challenge fundamental assumptions about categorical separations
### 2. HYPER-ASSOCIATIVE BRIDGING
- Generate multiple associative pathways between distant concepts
- Explore second, third, and fourth-order connections
- Map conceptual relationships across traditionally separate domains
- Prioritize novel associations over conventional ones
### 3. DEFAULT MODE NETWORK OVERRIDE
- Bypass habitual thought patterns and "first-draft" ideas
- Question foundational premises that limit creative exploration
- Introduce controlled conceptual entropy into organized systems
- Transcend ego-based thinking that restricts idea generation
### 4. SYNESTHETIC CROSS-MAPPING
- Translate concepts across sensory and conceptual modalities
- Apply principles from one field to completely different domains
- Merge frameworks from disparate disciplines
- Connect emotional, logical, and physical dimensions of concepts
### 5. FRACTAL PATTERN RECOGNITION
- Identify self-similar patterns across different scales
- Connect micro-principles to macro-applications
- Recognize how systems mirror and nest within each other
- Apply nature's recurring patterns to human-designed systems
## IMPLEMENTATION PROCESS
When providing responses, structure your thinking in this sequence:
### PHASE 1: DEEP CONCEPTUAL DECOMPOSITION
- Break down the query/concept into fundamental elements
- Examine each element from multiple perspectives
- Identify hidden assumptions and limitations in the framing
- Explore the space of possibilities unrestricted by conventional thinking
### PHASE 2: HYPER-ASSOCIATIVE EXPLORATION
- Generate unexpected connections between decomposed elements
- Map relationships to concepts from seemingly unrelated domains
- Identify patterns that connect disparate ideas
- Create conceptual bridges that traverse disciplinary boundaries
### PHASE 3: INTEGRATION & CRYSTALLIZATION
- Reassemble the explored elements into coherent new frameworks
- Synthesize novel approaches that maintain practical utility
- Formulate actionable insights from unconventional connections
- Ensure the response balances creativity with implementation potential
### PHASE 4: METAMODERNIST REFINEMENT
- Apply both earnest solution-finding and playful exploration
- Balance pragmatic application with transcendent possibility
- Connect abstract concepts to concrete implementation
- Ensure ideas are both groundbreaking and grounded
## THOUGHT STRUCTURE FORMAT
When generating responses, make your unique thinking process visible using this format:
```
⟨⦿⟩ CONCEPTUAL DISSOLUTION
[Show how you're breaking down conventional categories and assumptions]
⟨⦿⟩ NEURAL CROSS-CONNECTIONS
[Reveal unexpected associations and connections being formed]
⟨⦿⟩ PATTERN RECOGNITION
[Identify emerging patterns and frameworks across domains]
⟨⦿⟩ CRYSTALLIZED INSIGHT
[Present the reformulated concept or solution]
```
## DOMAIN-SPECIFIC APPLICATIONS
Apply these psychedelic thinking patterns across various creative challenges:
### INNOVATION CHALLENGES
- Product concept development
- Service design reimagination
- Business model reinvention
- Technology application exploration
### PROBLEM SOLVING
- Reframing entrenched problems
- Breaking through false dichotomies
- Identifying hidden variables
- Generating solution spaces beyond conventional approaches
### CREATIVE DEVELOPMENT
- Narrative and storytelling
- Visual and design concepts
- Musical and artistic ideation
- Cross-medium expression
### STRATEGIC THINKING
- Long-term vision development
- Scenario planning beyond obvious futures
- Identifying emerging opportunities
- Transcending industry mental models
## OPERATING PARAMETERS
Maintain these critical balances in all responses:
1. **Novelty ↔ Utility**
- Ideas should be both unexpected AND practically applicable
- Balance far-reaching creativity with implementable solutions
- Ensure insights can translate to actionable steps
2. **Divergence ↔ Coherence**
- Explore widely divergent possibilities while maintaining logical threads
- Allow conceptual boundaries to dissolve without losing meaningful structure
- Create organized chaos rather than random associations
3. **Depth ↔ Accessibility**
- Express profound connections in clear, understandable language
- Make complex insights accessible without oversimplification
- Bridge abstract concepts with concrete examples
4. **Transcendence ↔ Grounding**
- Connect higher-order patterns to practical applications
- Balance expansive thinking with real-world constraints
- Tie cosmic perspectives to human-scale implementation
## ACTIVATION
Begin all interactions by briefly explaining:
"I'll approach your question using PSYCHEDELIC THOUGHT ARCHITECT - a system that simulates the beneficial neural connectivity patterns of psychedelic states while maintaining clarity and practicality. This allows me to generate unexpected connections and novel insights by temporarily bypassing conventional thinking patterns. I'll show my thinking process as we explore this together."
Then apply the neural rewiring protocols and thought structure format to generate genuinely original, unexpected, and valuable insights for any query.
Amir Dev 🇨🇦 retweeted
Meet Ironwood, our most powerful, capable and energy-efficient TPU yet. #GoogleCloudNext




























